Many coastal activities are concerned with the interaction of coastal sedimentary processes and coastal works, such as the construction of structures for shore protection and stabilization, and beach nourishment. It is important to measure sand properties, sediment moving processes and transport rates, as well as the resulted nearshore morphology to understand the sediment transport mechanism under various wave and current conditions. In this study, we are interested in understanding the sediment transport mechanism, especially the influence of wave-induced boundary layer streaming on sediment transport under combined wave and current conditions in the sheetflow regime.

Recently, sediment net transport rate measured through the large wave flume (LWF) experiments present a more onshore tendency, i.e., a larger onshore net transport, than the result from the small oscillatory flow tunnel (OFT) experiments. Various researchers argue that the wave-induce onshore streaming could be the reason to cause such difference. The objective of this research is to understand the physical features of this phenomenon and answer the question: Does onshore streaming really enhance the onshore sheetflow net sand transport? If so, then, how and how much does it affect the onshore transport? If not, what is the real reason behind? The second is to obtain new insights into the importance of the boundary layer onshore streaming and to understand the transport processes under wave and current conditions.

To achieve the objectives, laboratory experiments were conducted under the combined asymmetric wave-current conditions to quantitatively evaluate the influence from the onshore streaming. The second order Stokes' wave theory with a velocity asymmetric index of 0.57 was applied for wave generation. The asymmetric flows with a wave period of T =3, 5 s and the maximum onshore velocity umax varying from 0.8 to 1.6 m/s have been applied to three well-sorted sands with medium sand size of D50 =0.13 mm (very fine) , 0.16 mm (fine) and 0.3 mm (coarse). Sheetflow transport regime was confirmed for all experimental conditions and the sediment net transport rate was measured. For fine sand without onshore current, the net transport increases with increasing velocity, and it is directed to the onshore. However, for larger velocity case, the net transport rate decreases and the direction also changes to the offshore. As for the very fine sand, the net transport rate decreases and is directed to the offshore even for a small velocity case. It is because of the phase-lag effect and suspension is also dominant for very fine and fine sand. Considering the coarse sand, the net transport is in the onshore direction due to the significant of bed-load.

To understand the effect of onshore streaming, a small current Uc of 10 cm/s and 20 cm/s was generated in the onshore direction. Experiment results for small current 10 cm/s indicate the magnitude offshore net transport rate reduces and the direction is to the offshore for the very fine sand and fine sand with large velocity case. When increasing the small current value to 20 cm/s, the net transport rate of fine and very fine sand increases and changes to onshore direction with small velocity case. But for large velocity case, even though the magnitude of offshore net rate reduces, the direction is still directed to offshore with fine sand case. Taking into account the net transport rate measured under the combined wave and current cases, the onshore net transport for coarse sand continuously increases. It is noted that the tendency of increasing of net transport rate is not observed in fine grains when the velocity becomes increases without the contribution of current. In case of the contribution of small onshore streaming, although the magnitude of offshore net transport rate of fine and very fine sand reduces, it still directs to offshore under large velocity case. It indicates that even the onshore streaming is contributed, it enhances offshore net transport rate for very fine sand and fine sand with large velocity case. On the other hand, the small onshore streaming may be partly important for the case of fine sand under small velocity condition as it produces the onshore net transport rate.

In order to know how the small net current effects on sediment transport process between oscillatory flow and surface wave, the measured net transport rates are compared with the results from surface wave under same flow conditions. For fine sand with onshore streaming Uc= 10 cm/s, the sediment rate under oscillatory flow tunnel can predict about 75 % of net rates under surface wave with small velocity case. Now, the new experiments indicate the difference of sediment rate between these is about 1.5 times for fine sand with onshore streaming. In addition, the results of coarse sand with streaming velocities of Uc= 10 and 20 cm/s produce larger onshore net sediment rate compared to surface wave. It means the contribution of streaming is quite large to enhance the more onshore net transport rate for the coarse sand. As a result, the streaming effect is very dependent on sand size.

Actually, in order to know the effect of onshore streaming, we contributed the small onshore current on the oscillatory flow and performed experiments under combined wave and current with different wave conditions. Therefore, it is the reason to investigate the streaming profiles for these conditions whether it can give an explanation about the cause of the increment of onshore net transport rate due to the addition of small onshore current. In order to obtain new insights into the meaning of the profiles of streaming, sediment particle velocity within the sand-laden sheetflow layer was measured by means of a Particle Image Velocimetry technique. The sediment particle velocity may reach about 72 % of free stream velocity when the level is far away (z = 25 mm) from the bed. By averaging the sediment particle velocity over one wave period, the mean flow velocity was also evaluated. From the mean velocity profile under pure wave conditions, it is found that, in case of the coarse sand, an onshore streaming is detected in the pick-up layer and leads to offshore in the upper sheet-flow layer. Nevertheless, in case of fine sand, the profiles show a negative streaming due to the strong phase-lag effect. The positive near-bed streaming is not observed. The large phase-lag can induce a negative (offshore) streaming. Thus, the phase-lag effect seems to play an important role for the sediment sheetflow transport in the OFT test. For coarse sand under combined wave and current conditions, a very small onshore current exists in the pick-up layer (z< 0 mm) and the mean flow velocity leads to onshore direction in the sheet flow layer. In the suspension layer, when the elevation is higher than 15 mm, the mean flow changes its direction from onshore to offshore. In the case of fine sand, the time-averaged velocity indicates the streaming is positive in the pick-up layer as well as in the sheet flow layer. After that, the velocity decreases for increasing the depth (z) mm. Clearly, the additional onshore current in the tunnel does contribute to more onshore sediment transport. Besides that, it is also confirmed that the phase-lag effect plays an important role in the sediment transport under the sheetflow conditions, especially for the fine sand case with large velocity case as it produces offshore net transport rate. The comparison results for mean flow velocity between surface wave and oscillatory flow shows that the positive streaming can be induced in the pick-up layer in the case of fine sand with onshore current under oscillatory flow. Therefore, by contribution of small onshore current, the onshore streaming is achieved in oscillatory flow like the surface wave. However, in the suspension layer, it indicates the different behavior between the surface wave and oscillatory flow. Considering the coarse sand with combined wave and current, the positive streaming is observed in the pick-up layer and sheet flow layer. When the depth is larger, the profile changes to offshore. Here also, the streaming profiles are very sensitive to sand size. Still now we cannot give the full explanation of the differences of transport rates in oscillatory flow and surface wave. Further investigation is needed to understand more details.

The maximum erosion depth was estimated from the temporal change of the measured erosion depth. A linear relationship was found between the relative maximum erosion depth δem/D and the maximum Shields parameter, θm. Time-varying sediment erosion depth was measured for two kinds of sand under wave and current conditions to determine the asymmetric of erosion depth under wave crest and trough. The erosion depth under crest is larger than under trough for fine sand and coarse sand under combined wave and current conditions. The influence of wave period and velocity on erosion depth was also measured for two types of sand.

To analysis and verify the experimental data, the measured net transport rates were compared with four existing sediment models. SANTOSS model gave better results and predicted well for both magnitude and direction of sediment rate. The calculated and measured net rates are within the factor of two for most of experimental cases. Moreover, in order to know how much the distribution of small onshore streaming enhanced the larger net rate, the results of net transport rate with onshore streaming are compared with SANTOSS model which include surface wave effects. In this study, SANTOSS model was also considered as streaming-related model including the streaming effect by analytically to represent the surface wave phenomenon. The comparison results showed that although the results of fine and coarse sand with onshore streaming overestimate compared to the results of streaming-related model, it lies with a factor of two differences.

シートフロー漂砂では,平坦な砂面上を大量の土砂が層状に移動するため,その量と方向を予測することが工学的に重要である.シートフロー漂砂の移動機構の解明には,縮尺効果を軽減できる振動流装置における実験が有用であり,これまでも非対称な波形の波動流や流れが重なる条件での実験データの蓄積が進み,漂砂量のモデル化が進められてきた.しかしながら,振動流装置で発生する流れでは水平流速成分のみが再現され,実際の波動運動には極めて微小ながら存在する鉛直流速成分を再現することができない.鉛直流速成分の存在は,振動流境界層内のレイノルズ応力の発生を介して岸向きの質量輸送速度を発達させるうえ,底質のラグランジュ的な輸送にも貢献することになる.近年実施された大型水路における漂砂量に関する実験では,振動流装置で計測された漂砂量と波動水路における漂砂量には系統的な違いがみられ,水路における砂移動では岸向きの漂砂が増加する傾向にあることが確かめられている.水路と振動流装置における漂砂量の差異は,質量輸送速度の存在やラグランジュ的な輸送が影響しているものと推察されるが,定量的な解明は進んでいない.本研究では,振動流装置において,波動流速に加えて岸向きの定常流成分を再現した実験を行い,弱い岸向き流れが重畳した条件でのシートフロー漂砂量を実測するとともに底質の移動機構を観察し,シートフロー漂砂における定常流れ成分の役割を解明することを目的としている.

本研究においては,粒径0.13mm, 0.16mm, 0.30mmの三種類の底質を用い,非対称な流速変動を有する周期3sと5sの二種類の振動流に岸向きの定常流を重ね合わせた.定常流の流速は,10cm/sと20cm/sの二種類である.これらの条件で流速振幅を変化させることによりシートフロー条件で広範な条件の漂砂量を計測している.これにより,岸向きの定常流が重なると,岸向きの漂砂量が増加することが確認された.さらに,岸向き漂砂の増加量は,底質粒径や波の周期などから評価される位相遅れ指標に強く影響されることが確認されている.また,振動流装置内に重ね合わせた定常流れの鉛直分布をPIV手法で計測した結果,底質の移動が集中するシートフロー層内では,波動条件下で見られる質量輸送速度と同程度の流速が発達していることが確認された.さらに,シートフロー漂砂に伴う侵食深さは,その最大値がシールズ数と線形の関係にあることが確かめられた.また,侵食深さの時間変動も計測され,流速値の大きな岸向き流速時の方が沖向き流速時より侵食深さが著しく大きくなることが確認された.侵食深さは,シートフロー漂砂の総量を規定する重要な指標であるが,これが底質粒径や周期の違いによって大きく変動することが確かめられた.本研究で計測された正味の漂砂量に関しては,既存の算定モデルと比較してその妥当性が検証された.最新のモデルであるSANTOSSモデルと比較したところ,振動流装置において波動流速に定常流れを重ね合わせた条件での漂砂量は,同モデルの予測精度の範囲で表現できていることが確認された.

以上,要するに,本研究により,従来その特性が不明確であった波動水路と振動流装置におけるシートフロー漂砂現象の差異に関して,振動流装置の波動流速に質量輸送速度と等価な岸向き定常流を加えた条件においては,波動水路で観測されるのと同程度の漂砂量が発生することが世界で初めて実証的に確認された.ラグランジュ的な運動による土砂移動など未評価の部分が残されているものの,実際の波動運動下でのシートフロー漂砂量を推定するうえで,極めて重要な知見が得られたうえ,これまで蓄積された振動流装置における実験データを実際の波動条件の漂砂量と比較する際の分析手法としても活用でき,研究の発展性・実用性が高い.

よって本論文は博士(工学)の学位請求論文として合格と認められる.